Hey — we built VARRD because we were sick of seeing people get sold overfit backtests and fake
signals. The trading education space is full of garbage: someone tests 500 indicator combos,
shows you the one that hit, and calls it a strategy.
VARRD lets you take any trading idea — even a vague one — and put it through the same
statistical validation process institutional quant desks use. You just talk to it in plain
English. It does the research, builds the pattern, and tests it for you.
The key thing is it's fully guardrailed against cheating its own results:
Tracks every hypothesis tested (K-tracking) so you can't cherry-pick winners
Out-of-sample data is used exactly once per hypothesis — no peeking, no re-running
AI can't fabricate stats — every number comes from executed code against real data
Proper event study methodology, not just "did the backtest make money"
It runs autonomously and averages about 3-4 statistically valid edges per hour. It shows you
exactly when they fire and the entry/exit points.
We're in early access and only charging enough to cover compute costs. The goal is to get this
in front of as many people as possible — ideally through a brokerage acquisition — so regular
people stop getting scammed by people selling lines on a chart.
Hey — we built VARRD because we were sick of seeing people get sold overfit backtests and fake signals. The trading education space is full of garbage: someone tests 500 indicator combos,
shows you the one that hit, and calls it a strategy.
VARRD lets you take any trading idea — even a vague one — and put it through the same
statistical validation process institutional quant desks use. You just talk to it in plain English. It does the research, builds the pattern, and tests it for you.
The key thing is it's fully guardrailed against cheating its own results:
It runs autonomously and averages about 3-4 statistically valid edges per hour. It shows you exactly when they fire and the entry/exit points.
We're in early access and only charging enough to cover compute costs. The goal is to get this in front of as many people as possible — ideally through a brokerage acquisition — so regular people stop getting scammed by people selling lines on a chart.